Snowdon Pharmaceuticals, a three-year-old biotech based at 1 Deer Park Drive, has three areas of focus — pain, infectious disease, and neurological disorders. And the thread that holds together these seemingly wide-ranging research areas is a technique for discovering potential drug candidates.
The company’s founder is William Welsh, a professor of bioinformatics and computer-aided molecular design in the department of pharmacology at the Robert Wood Johnson Medical School and the University of Medicine and Dentistry of New Jersey. Welsh developed Snowdon’s computer-based model to accelerate the front end of the drug discovery process. “By accelerating the process, you also make it less expensive,” says Welsh. “Much of the expense of drug discovery is spent in the time you have to search through and find the correct molecule that has the properties that would make for a safe and effective drug in humans.”
Welsh’s software is a form of virtual screening that allows scientists to visualize molecules on the screen and investigate the properties of potential drug molecules that allow them to bind and work inside the human body as desired. The way these molecules work is that by binding to a protein in your body, they change the course of the illness or disease.
What differentiates Snowdon Pharmaceuticals, he says, is that it converts each of the approximately 5 million potential molecules in its database into an electronic “signature” that encodes key information about the molecule’s shape and size as well as surface features like the distribution of positive and negative charges. “We don’t have actual samples,” he says. “We have images on a computer screen.”
So far federal and state government agencies have seen enough promise in Snowdon to be the company’s main source of income. Snowdon has received a two-year grant for $628,000 from the U.S. National Institutes of Health; a contract with the U.S. Food & Drug Administration to develop computational tools for rapid detection of contaminants in commercial pharmaceutical products; a $114,000 grant from the NIH to support the company's development of novel treatments for colorectal cancer; and a two-year $500,000 Edison Innovation Award grant from the New Jersey Commission on Science & Technology to develop medicinal products for the treatment of toxoplasmosis and related serious parasitic infections.
Welsh says the grants have stimulated partnerships with biopharmaceutical companies to accelerate the clinical and commercial development of these potential therapeutic agents.
In a sifting process, Snowdon locates potential drug candidates that will bind to the disease protein of interest. “We are looking for those molecules that will fit like a hand into a glove or a key into a keyhole into the protein’s binding pocket,” he says.
Welsh’s computer program is able to compare the shape of each drug molecule with the inside shape of the pocket. Then the molecules are ranked according to how well they fit.
After determining whether the shapes of the small molecule and the protein pocket are complementary, the program compares their electrical properties. Molecules and proteins have positive and negative charges. “When you are trying to fit molecules into a pocket, where the molecule has a positive part, you want the protein pocket to have a negative charge, because opposite charges attract,” he says.
Shape usually narrows down the number of molecules to a couple thousand, and then charges narrow it to a couple hundred.
The computer then docks each drug molecule individually into the protein pocket and scores them from 0 to 100, based on how good the fit is. “Regardless of what the disease is, we can sift through molecules,” says Welsh. “The only thing changing is the protein.”
Now down to the top 50 to 100 molecules, they purchase small quantities of maybe the fifth top-ranked molecules, less than a gram, and do some biological studies to see whether the molecules work in the way the computer predicted they would.
For the molecules that have made it through the tests thus far, Snowdon’s chemists inspect them again on the computer screen and think about how they can change its structure in ways that will make it safer and more effective.
“Since we are already starting with a molecule that is active, we are off to a jumpstart,” says Welsh. The result of this process, he says, is finding “an enriched set of molecules that have a higher probability to lead to successful drugs.”
As a professor working in academic laboratories, Welsh developed expertise on a protein involved in neuropathic pain — the kind that does not go away and is associated with conditions like shingles, diabetes, AIDS, fibromyalgia, and restless leg syndrome. No cures exist for these types of pain, and the few treatments that are available have negative side effects, says Welsh.
“We thought this would be an unmet need. As we studied the protein, we found out that it is involved in many other conditions and neurological diseases, like schizophrenia, bipolar disorder, depression, anxiety, and autism,” he says. This single protein, therefore, has led to work on a drug for treating neuropathic pain and others for treating neurological disease. Neuropathic pain is the company’s most advanced program.
One condition for which Snowdon is developing a drug is fragile x, a serious form of autism. “The protein we are studying is directly associated with the development of fragile x and aggravation of the disease,” he explains. “What happens in fragile x is that the absence of a certain protein causes all the symptoms, and the molecule we are developing would indirectly replenish that protein.” In animal studies, he says, the drug candidate has essentially eliminated the disease’s symptoms, but right now Snowdon does not have the funding to move ahead quickly. “Fragile x is a fairly rare disease;” he says, “and the rarer a disease is, the more difficult it is to get financial support to develop treatments.”
Snowdon has also been working on various infectious diseases, like tuberculosis, in collaboration with Nancy Connell, professor in the department of medicine at UMDNJ in Newark. With Connell, Snowdon applied to the National Institutes of Health to work on tuberculosis, which is a deadly global disease that is infecting more and more American citizens — and which scientists are having difficulty treating.
This work in tuberculosis led to work in another, related direction, also with Connell, on biowarfare pathogens for the Department of Defense. What was attractive to the department was Snowdon’s computational strategy, which could be used to find novel drugs to combat biowarfare pathogens — bacteria that an enemy would use on either soldiers in the field or to provoke a homeland security crisis.
After expanding its application based on the department’s needs, Snowdon received an $8.2 million contract to use its method for finding drugs to combat particular bacteria, like anthrax and bubonic plague. “These are very nasty things, and they could be carried or delivered in ways that could contaminate and kill large numbers of people very rapidly,” Welsh says.
Welsh’s parents came from humble backgrounds. Neither went beyond eighth grade. “I was the first one in the family to go to college,” says Welsh. But his parents’ humble example paved a route for him in the business world. “I think that what I call the ‘common touch’ is a very important part of me,” he says. “You can come from a humble background or nobility and you’re still the same person, and you should be treated accordingly.” When hiring he looks for that same “common touch.”
Although Welsh’s true academic loves are history and psychology, he got interested in science in high school in Philadelphia, where he grew up. After graduating from St. Joseph’s University in 1969 with a bachelor’s in chemistry he earned his Ph.D. in theoretical physical chemistry from Penn in 1975.
He then took a position with Proctor and Gamble in Cincinnati, working for four years in applied research on products to clean floors and toilet bowls. Because he had always wanted to end up in academe, he became a postdoctoral research associate at the University of Cincinnati and also spent six months at the National Institutes of Health.
This was during a period when computers were becoming more widely available and scientists started to use computers in their research. At the University of Cincinnati, scientists were using computers to visualize what molecules actually looked like, which Welsh calls “the birth of a new technology.” They pulled him into their research, with the idea that he would use his computational knowledge to help them make drug molecules.
It was a bootstrap operation where he learned what he needed to know on his own; and he started to work on developing drug molecules and visualizing how they interact with proteins. Although the field has changed over time, with the cracking of the genome, faster and less expensive computers, and sophisticated software, he has stayed put, noting that “there are plenty of bright people to collaborate and interact with.”
After six years in Cincinnati he became the director of the Laboratory for Computer-Aided Molecular Design at the University of Missouri in St. Louis, where he stayed for 15 years. In 2001, when his wife’s company was acquired and she had to relocate to the east coast to work at Johnson & Johnson, Welsh moved with her to New Jersey.
When it came to naming the company, Welsh wanted a name that was not technical and had nice associations. The name derives from Snowden Lane in Princeton, near Welsh’s home, and from Snowdonia, the highest mountain in Wales. It was a family vacation spot for his wife, who is English.
Currently Snowdon Pharmaceuticals has about 20 people, which includes part-time employees and contractors. When he hires, Welsh always tries to make sure that people fit both into the culture at Snowdon and with each other.
Snowdon Pharmaceuticals, 1 Deer Park Drive, Suite H-3, Monmouth Junctions 08852; 609-430-1957. Bill and Susan Welsh, founders. www.snowdonpharma.com